Association of antithrombin with development of trauma-induced disseminated intravascular coagulation and outcomes.

Abstract

Trauma activates the innate immune system to modulate hemostasis and minimize the damage caused by physiological bodily responses, including the activation of coagulation. Sufficiently severe trauma overwhelms physiological responses and elicits the systemic inflammatory response syndrome, which leads to the onset of disseminated intravascular coagulation (DIC), characterized by dysregulated inflammatory coagulofibrinolytic responses. Impaired anticoagulant mechanisms, including antithrombin, constitutes the pathology of DIC, while the dynamics of antithrombin and relevance to outcomes in trauma-induced coagulopathy have not been fully elucidated. This study investigated the associations of antithrombin activity with DIC onset and outcomes in severely injured patients. This retrospective sub-analysis of a multicenter, prospective study included patients with an injury severity score ≥16. We characterized trauma patients with low antithrombin activity (antithrombin <80% on hospital arrival, n = 75) in comparison with those who had normal antithrombin activity (antithrombin ≥80%, n = 200). Global markers of coagulation and fibrinolysis, molecular biomarkers for thrombin generation (soluble fibrin [SF]), and markers of anticoagulation (antithrombin) were evaluated to confirm the associations of antithrombin with DIC development and outcomes, including in-hospital mortality and the multiple organ dysfunction syndrome (MODS). Patients with low antithrombin activity had higher prevalence of shock, transfusion requirements, and in-hospital mortality. Higher DIC scores and more severe organ dysfunction were observed in the low AT group compared to that in the normal AT group. Antithrombin activity on arrival at the hospital was an independent predictor of the development of DIC in trauma patients, and levels of SF increased with lower antithrombin values (antithrombin activity > 85%). Antithrombin activity at 3 h showed good predictive performance for in-hospital mortality, and a multivariable Cox proportional-hazard regression model with a cross-product term between the antithrombin and DIC showed that the in-hospital mortality in patients with DIC increased with decreased antithrombin activity. A multivariable logistic regression model showed that the odds for the development of MODS in patients with DIC increased with lower antithrombin values. Decreased antithrombin activity in trauma-induced coagulopathy is associated with poor outcomes through worsening of DIC.